Inpatient Care

Einleitung / Introduction

In numerous countries, policy uses training subsidies in order to increase the supply of firm-provided training. The same holds for levy-grant schemes that redistribute apprenticeship costs between firms that provide training and firms that do not provide training. However, despite of the wide-reaching usage of training subsidies and levies by policy, empirical evidence on the causal effects of such instruments on training supply of firms is barely available. This paper tackles this research question and implements causal inference into this research question by considering the specific situation of care workers in Germany. To stimulate incentives for care facilities to provide training, the German federal government has created the possibility of introducing an obligatory training levy that finances the apprenticeship of geriatric nurses in inpatient and ambulatory care. As a consequence, half of the federal states introduced such a levy that redistributes the apprentice's pay of about €17,000 per apprentice per year between facilities that train and facilities that do not train.

Methode / Method

I use this exogenous variation in apprenticeship costs and take advantage of the fact that this substantial redistribution of costs was introduced at different points in time across the federal states. By using panel data of the Establishment History Panel (BHP), this enables me to estimate effects on the apprenticeship supply of care facilities. The variation in time and region solves the problem that in related studies an appropriate control group is often missing because levies are usually introduced country-wide and in industries with labor supply shortage. Next to this methodological contribution, the quasi-experimental design provides several contributions for more general interest.

Ergebnisse / Results

For ambulatory nursing services, where initial training supply before the introduction of the levy is rather low, I find an immediate robust positive effect on the probability to train apprentices and on the number of apprentices. For inpatient care facilities, immediate effects are absent. However, by accounting for the temporal evolution of treatment effects, I demonstrate that in the years after the enactment of the levy, effects for both ambulatory and inpatient facilities are significantly positive for the probability to train and for the quantity of training new apprentices. Further analyses uncover that effects are stronger in areas where the demand for care workers is the largest. Moreover, I provide robustness checks to show that those effects are due to the introduction of the levy scheme and demonstrate that the effects are not attributable to other reforms regarding the care sector. Moreover, the effects are not driven by endogenous market entry and exit.

AutorInnen

Eric Schuss, Bundesinstitut für Berufsbildung (BIBB)

Einleitung / Introduction

Achieving high-quality care is the paramount goal of any health care system. In the context of hospitals, there are two measures to ensure high quality: system certification and innovative activities. While the former aims at a high degree of standardization to reduce variation and errors in processes, the latter requires a high degree of flexibility to develop and introduce new technologies and processes. Given the scarcity of resources and partly contradicting requirements, it is unclear how system certification and innovative activities influence each other. However, such an understanding is needed to make strategic decisions to allocate resources when approaching certification and innovative activities simultaneously, while also being able to take appropriate measures in case of contrary effects. Taking organ cancer centers in Germany as an example, the aim of this study is to explore the causal relationship between certification activities and innovativeness and develop a structural model including environmental and organizational determinants.

Methode / Method

We conducted semi-structured interviews including 11 clinicians and 14 quality managers across 20 certified centers between June and September 2020. In a grounded theory inductive approach, we apply the Gioia-method to code the interview contents across multiple waves for a qualitative analysis. We first collected individual statements which were aggregated to overarching themes and finally to broader dimensions. Based on an in-depth analysis of the coded content structures, we identified causal paths of these dimensions to form the structural model.

Ergebnisse / Results

Our results suggest that the intention to be certified is highly dependent on the market context as well as the strategic goals of a hospital. For smaller hospitals, certification is a sufficient condition to stay competitive in the market. Due to causing standardization, certification has a positive effect on incremental process and product innovation. For larger hospitals, obtaining certification is a necessary condition to acquire additional funding, but the influence on innovative activity is limited due to the fact that the level of innovation is already high.

Zusammenfassung / Conclusion

Whether certification interferes with innovative activity is perceived differently across market contexts. Seeking system certification does not uniformly change innovative activity. Understanding this relationship can help hospital managers and policy makers to decide when to allocate resources towards either approach.

AutorInnen

Tim Brand, Universität Duisburg-Essen

Katharina Blankart, Universität Duisburg-Essen

Einleitung / Introduction

Medical guidelines aim to promote patient health and control of healthcare costs by setting evidence-based standards of care, thereby simplifying medical decision-making and addressing variations in medical practice. This study builds upon the existing literature by exploring spatial dependencies and other influencing factors in the adoption of drug-eluting stents (DES) in Germany and Italy, two of the largest medical device markets in Europe. We specifically investigate spillover effects in 2012-2016, after the European Society of Cardiology issued a guideline generally recommending the use of DES in 2012.

Methode / Method

We base our analysis on administrative data of patients with ST-elevation myocardial infarction (STEMI), who were admitted to a hospital in 2012-2016. Our dependent variable is the utilization of DES to total stents in patients with a STEMI diagnosis. In addition, we control for relevant hospital and regional factors. First, we test for the presence of spatial autocorrelation in the outcome variable using the Moran’s I test. Subsequently, we estimate spatial panel models allowing for global spillover effects. The relative contribution of geographic proximity between a hospital and its peers is examined by applying three different spatial weight matrices based on inverse distances between hospitals: (i) distance to neighboring hospitals, (ii) distance to neighboring university hospitals, and (iii) distance to neighboring private hospitals.

Ergebnisse / Results

We find significant positive spatial autocorrelation between neighboring hospitals in Germany in each year based on the Moran’s I test. In addition, the spatial lag parameter is significant and positive across different model specifications and weight matrices. In other words, for the case of DES in Germany, we find that a given hospital responds to the guideline uptake of its peers. In addition, we find positive spillover effects of university hospitals on non-university hospitals and of private hospitals on public/non-profit hospitals, however not consistently for all years.

Zusammenfassung / Conclusion

First results for Germany indicate the presence of positive spatial spillovers between peers regarding the uptake of medical guidelines for the case of DES e.g., through knowledge transfers or organizational learning. Further, the results suggest that university and private hospitals were "early adopters" of DES and that the spillover effects originated at least partially from these hospital types. Preliminary results for Italy at the district level have shown important differences in terms of spillover effects compared to Germany. In the further course of this project, the analyses in Italy will be conducted at the hospital-level to enable more concrete statements on the differences between the two countries and discuss possible explanations.

AutorInnen

Meilin Möllenkamp, Hamburg Center for Health Economics, University of Hamburg

Benedetta Pongiglione, Centre for Research on Health and Social Care Management, Bocconi University

Stefan Rabbe, Hamburg Center for Health Economics, University of Hamburg

Aleksandra Torbica, Centre for Research on Health and Social Care Management, Bocconi University

Jonas Schreyögg, Hamburg Center for Health Economics, University of Hamburg

Einleitung / Introduction

A significant fraction of unplanned hospital admissions are hospitalizations for Ambulatory Care Sensitive Conditions (ACSCs). Those hospitalizations could in many cases be avoided through the timely, ongoing application of preventive care or early disease management in the ambulatory setting. Health policy decision-makers achieve allocative efficiency if resources directed to the health and social sectors reach equal marginal benefit. Despite this, many health systems still focus strongly on health care interventions in order to reduce hospital admissions, although it may be more efficient to implement social care measures. Our systematic review provides evidence to what extent cooperation of the health care sector with the social care sector can effectively and efficiently reduce ACSC hospitalizations. We synthesize evidence on the configuration and coordination-related determinants of cross-sectoral partnerships between the health care and social care sector that influence ACSC hospitalization rates. Therefore, we evaluate (1) the composition and structure of cross-sectoral partnerships and (2) the formal care coordination processes.

Methode / Method

We conducted a systematic review of peer-reviewed literature from Medline, CINAHL, Embase and Scopus databases in June 2020. After screening title and abstract of 1106 articles, we included 27 in the final analysis.

Ergebnisse / Results

The most commonly implemented coordination activities are supporting patients’ self-management goals, establishing accountability or negotiating responsibility and facilitating transitions as coordination needs change as well as the broader concept of care coordination. The number and diversity of actors involved varied strongly between interventions. We found that most interventions are characterized by high interconnectedness of actors, but that the contribution of the non-health sector actors is less strong than the contribution of the health care actor. We could not identify defined architypes of collaborative interventions that determine success or failure in reducing hospitalizations.

Zusammenfassung / Conclusion

Our results inform policy makers about the effects of cross-sectoral interventions for reducing ACSC hospitalizations. With regard to configuration, all successful interventions were characterized by high interconnectedness. However, we could neither find evidence that the sheer number of care coordination activities nor that the configuration of a collaboration and its attributes of care coordination offered increases effectiveness of an intervention.

AutorInnen

Vanessa Ress, Hamburg Center for Health Economics (HCHE)

Lize Duminy, Universität Bern

Eva-Maria Wild, Hamburg Center for Health Economics (HCHE)

Rudolf Blankart, Universität Bern